Digital Communications and Networks最新文献

Jamming recognition method based on wavelet packet decomposition and improved deep learning 基于小波包分解和改进深度学习的干扰识别方法

IF 7.5 2区计算机科学 Q1 TELECOMMUNICATIONS

Digital Communications and Networks

Pub Date : 2025-10-01 DOI: 10.1016/j.dcan.2025.05.004

Qi Wu , Gang Li , Xiang Wang , Hao Luo , Lianghong Li , Qianbin Chen , Xiaorong Jing

To overcome the challenges of poor real-time performance, limited scalability, and low intelligence in conventional jamming pattern recognition methods, this paper proposes a method based on Wavelet Packet Decomposition (WPD) and enhanced deep learning techniques. In the proposed method, an agent at the receiver processes the received signal using WPD to generate an initial Spectrogram Waterfall (SW), which is subsequently segmented using a sliding window to serve as the input for the jamming recognition network. The network employs a bilateral filter to preprocess the input SW, thereby enhancing the edge features of the jamming signals. To extract abstract features, depthwise separable convolution is utilized instead of traditional convolution, thereby reducing the network's parameter count and enhancing real-time performance. A pyramid pooling layer is integrated before the fully connected layer to enable the network to process input SW of varying sizes, thus enhancing scalability. During network training, adaptive moment estimation is employed as the optimizer, allowing the network to dynamically adjust the learning rate and accelerate convergence. A comprehensive comparison between the proposed jamming recognition network and six other models is conducted, along with Ablation Experiments (AE) based on numerical simulations. Simulation results demonstrate that the proposed method based on WPD and enhanced deep learning achieves high-precision recognition of various jamming patterns while maintaining a favorable balance among prediction accuracy, network complexity, and prediction time.

针对传统干扰模式识别方法实时性差、可扩展性有限、智能程度低等问题，提出了一种基于小波包分解（WPD）和增强深度学习技术的干扰模式识别方法。在该方法中，接收端的代理使用WPD对接收信号进行处理，生成初始谱图瀑布（SW），随后使用滑动窗口对其进行分割，作为干扰识别网络的输入。该网络采用双边滤波器对输入的SW进行预处理，从而增强了干扰信号的边缘特征。为了提取抽象特征，采用深度可分离卷积代替传统的卷积，从而减少了网络的参数个数，提高了实时性。在全连接层之前集成了金字塔池层，使网络能够处理不同大小的输入SW，从而增强了可扩展性。在网络训练过程中，采用自适应矩估计作为优化器，使网络能够动态调整学习率，加快收敛速度。对所提出的干扰识别网络与其他六种模型进行了全面比较，并进行了基于数值模拟的消蚀实验。仿真结果表明，基于WPD和增强深度学习的方法在保持预测精度、网络复杂度和预测时间的良好平衡的同时，实现了对各种干扰模式的高精度识别。

{"title":"Jamming recognition method based on wavelet packet decomposition and improved deep learning","authors":"Qi Wu , Gang Li , Xiang Wang , Hao Luo , Lianghong Li , Qianbin Chen , Xiaorong Jing","doi":"10.1016/j.dcan.2025.05.004","DOIUrl":"10.1016/j.dcan.2025.05.004","url":null,"abstract":"<div><div>To overcome the challenges of poor real-time performance, limited scalability, and low intelligence in conventional jamming pattern recognition methods, this paper proposes a method based on Wavelet Packet Decomposition (WPD) and enhanced deep learning techniques. In the proposed method, an agent at the receiver processes the received signal using WPD to generate an initial Spectrogram Waterfall (SW), which is subsequently segmented using a sliding window to serve as the input for the jamming recognition network. The network employs a bilateral filter to preprocess the input SW, thereby enhancing the edge features of the jamming signals. To extract abstract features, depthwise separable convolution is utilized instead of traditional convolution, thereby reducing the network's parameter count and enhancing real-time performance. A pyramid pooling layer is integrated before the fully connected layer to enable the network to process input SW of varying sizes, thus enhancing scalability. During network training, adaptive moment estimation is employed as the optimizer, allowing the network to dynamically adjust the learning rate and accelerate convergence. A comprehensive comparison between the proposed jamming recognition network and six other models is conducted, along with Ablation Experiments (AE) based on numerical simulations. Simulation results demonstrate that the proposed method based on WPD and enhanced deep learning achieves high-precision recognition of various jamming patterns while maintaining a favorable balance among prediction accuracy, network complexity, and prediction time.</div></div>","PeriodicalId":48631,"journal":{"name":"Digital Communications and Networks","volume":"11 5","pages":"Pages 1469-1478"},"PeriodicalIF":7.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529005","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Building robust traffic classifier under low quality data: A federated contrastive learning approach 在低质量数据下构建健壮的流量分类器：一种联邦对比学习方法

IF 7.5 2区计算机科学 Q1 TELECOMMUNICATIONS

Digital Communications and Networks

Pub Date : 2025-10-01 DOI: 10.1016/j.dcan.2025.06.001

Tian Qin , Guang Cheng , Zhichao Yin , Yichen Wei , Zifan Yao , Zihan Chen

In the big data era, the surge in network traffic volume poses challenges for network management and cybersecurity. Network Traffic Classification (NTC) employs deep learning to categorize traffic data, aiding security and analysis systems as Intrusion Detection Systems (IDS) and Intrusion Prevention Systems (IPS). However, current NTC methods, based on isolated network simulations, usually fail to adapt to new protocols and applications and ignore the effects of network conditions and user behavior on traffic patterns. To improve network traffic management insights, federated learning frameworks have been proposed to aggregate diverse traffic data for collaborative model training. This approach faces challenges like data integrity, label noise, packet loss, and skewed data distributions. While label noise can be mitigated through the use of sophisticated traffic labeling tools, other issues such as packet loss and skewed data distributions encountered in Network Packet Brokers (NPB) can severely impede the efficacy of federated learning algorithms. In this paper, we introduced the Robust Traffic Classifier with Federated Contrastive Learning (FC-RTC), combining federated and contrastive learning methods. Using the Supcon-Loss function from contrastive learning, FC-RTC distinguishes between similar and dissimilar samples. Training by sample pairs, FC-RTC effectively updates when receiving corrupted traffic data with packet loss or disorder. In cases of sample imbalance, contrastive loss functions for similar samples reduce model bias towards higher proportion data. By addressing uneven data distribution and packet loss, our system enhances its capability to adapt and perform accurately in real-world network traffic analysis, meeting the specific demands of this complex field.

在大数据时代，网络流量的激增对网络管理和网络安全提出了挑战。NTC （Network Traffic Classification）是一种利用深度学习对流量数据进行分类的技术，为安全分析系统提供入侵检测系统（IDS）和入侵防御系统（IPS）的支持。然而，目前的NTC方法基于孤立的网络模拟，通常不能适应新的协议和应用，并且忽略了网络条件和用户行为对流量模式的影响。为了提高网络流量管理的洞察力，已经提出了联邦学习框架来聚合不同的流量数据以进行协作模型训练。这种方法面临着数据完整性、标签噪声、数据包丢失和倾斜数据分布等挑战。虽然可以通过使用复杂的流量标记工具来减轻标签噪声，但网络数据包代理（NPB）中遇到的数据包丢失和数据分布偏斜等其他问题可能严重阻碍联邦学习算法的有效性。本文介绍了联邦对比学习鲁棒流量分类器（FC-RTC），它将联邦学习和对比学习相结合。使用来自对比学习的Supcon-Loss函数，FC-RTC区分相似和不相似的样本。通过样本对训练，FC-RTC在接收到丢失或无序的损坏流量数据时有效地进行更新。在样本不平衡的情况下，相似样本的对比损失函数减少了模型对高比例数据的偏差。通过解决数据分布不均和丢包问题，提高了系统在实际网络流量分析中的适应性和准确性，满足了这一复杂领域的具体需求。

{"title":"Building robust traffic classifier under low quality data: A federated contrastive learning approach","authors":"Tian Qin , Guang Cheng , Zhichao Yin , Yichen Wei , Zifan Yao , Zihan Chen","doi":"10.1016/j.dcan.2025.06.001","DOIUrl":"10.1016/j.dcan.2025.06.001","url":null,"abstract":"<div><div>In the big data era, the surge in network traffic volume poses challenges for network management and cybersecurity. Network Traffic Classification (NTC) employs deep learning to categorize traffic data, aiding security and analysis systems as Intrusion Detection Systems (IDS) and Intrusion Prevention Systems (IPS). However, current NTC methods, based on isolated network simulations, usually fail to adapt to new protocols and applications and ignore the effects of network conditions and user behavior on traffic patterns. To improve network traffic management insights, federated learning frameworks have been proposed to aggregate diverse traffic data for collaborative model training. This approach faces challenges like data integrity, label noise, packet loss, and skewed data distributions. While label noise can be mitigated through the use of sophisticated traffic labeling tools, other issues such as packet loss and skewed data distributions encountered in Network Packet Brokers (NPB) can severely impede the efficacy of federated learning algorithms. In this paper, we introduced the Robust Traffic Classifier with Federated Contrastive Learning (FC-RTC), combining federated and contrastive learning methods. Using the Supcon-Loss function from contrastive learning, FC-RTC distinguishes between similar and dissimilar samples. Training by sample pairs, FC-RTC effectively updates when receiving corrupted traffic data with packet loss or disorder. In cases of sample imbalance, contrastive loss functions for similar samples reduce model bias towards higher proportion data. By addressing uneven data distribution and packet loss, our system enhances its capability to adapt and perform accurately in real-world network traffic analysis, meeting the specific demands of this complex field.</div></div>","PeriodicalId":48631,"journal":{"name":"Digital Communications and Networks","volume":"11 5","pages":"Pages 1479-1492"},"PeriodicalIF":7.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529379","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

CARS: Connection as required scheme for horizontal communications in Industry 4.0 CARS：工业4.0水平通信的按需连接方案

IF 7.5 2区计算机科学 Q1 TELECOMMUNICATIONS

Digital Communications and Networks

Pub Date : 2025-10-01 DOI: 10.1016/j.dcan.2025.03.003

Jianhua Li , Bohao Feng , Aleteng Tian , Hui Zheng , Klaus Moessner , Hong-ning Dai , Jiong Jin

In the rapidly evolving landscape of Industry 4.0 (I4.0), the convergence of information and operational technologies necessitates real-time communication and collaboration across cyber-physical systems and the Internet of Things (IoT). Rapid data transmission is particularly critical within enterprises (vertically) and among stakeholders (horizontally) in this complex, heterogeneous ecosystem. While current research has focused on data application, processing, and storage within the cloud-edge-device continuum, cross-edge transmission has received less attention, resulting in challenges such as suboptimal routing and excessive delays in horizontal communications. To address the above issues, this paper introduces a Connection-As-Required Scheme (CARS) specifically designed for delay-sensitive IoT and Cyber-Physical System (CPS) applications, where low-latency communication is essential for operational efficiency. CARS leverages Lyapunov optimization and backpressure algorithms to optimize traffic scheduling and routing, minimizing communication delay between entities. Benchmarking against state-of-the-art solutions, CARS reduces Round-Trip Time (RTT) to approximately 47.0% of conventional methods and decreases delay by 24.5% in TCP-based and 26.0% in UDP-based applications. These results highlight the potential of CARS to facilitate effective, low-latency collaboration in diverse I4.0 environments.

在快速发展的工业4.0 （I4.0）环境中，信息和操作技术的融合需要跨网络物理系统和物联网（IoT）的实时通信和协作。在这个复杂、异构的生态系统中，快速数据传输在企业内部（垂直）和利益相关者之间（水平）尤为重要。虽然目前的研究主要集中在云边缘设备连续体中的数据应用、处理和存储，但交叉边缘传输受到的关注较少，这导致了水平通信中的次优路由和过度延迟等挑战。为了解决上述问题，本文介绍了专为延迟敏感物联网和网络物理系统（CPS）应用而设计的按需连接方案（CARS），其中低延迟通信对于运营效率至关重要。CARS利用Lyapunov优化和背压算法来优化交通调度和路由，最大限度地减少实体之间的通信延迟。与最先进的解决方案相比，CARS将往返时间（RTT）减少到传统方法的47.0%左右，在基于tcp的应用中减少24.5%，在基于udp的应用中减少26.0%的延迟。这些结果突出了CARS在各种工业4.0环境中促进有效、低延迟协作的潜力。

{"title":"CARS: Connection as required scheme for horizontal communications in Industry 4.0","authors":"Jianhua Li , Bohao Feng , Aleteng Tian , Hui Zheng , Klaus Moessner , Hong-ning Dai , Jiong Jin","doi":"10.1016/j.dcan.2025.03.003","DOIUrl":"10.1016/j.dcan.2025.03.003","url":null,"abstract":"<div><div>In the rapidly evolving landscape of Industry 4.0 (I4.0), the convergence of information and operational technologies necessitates real-time communication and collaboration across cyber-physical systems and the Internet of Things (IoT). Rapid data transmission is particularly critical within enterprises (vertically) and among stakeholders (horizontally) in this complex, heterogeneous ecosystem. While current research has focused on data application, processing, and storage within the cloud-edge-device continuum, cross-edge transmission has received less attention, resulting in challenges such as suboptimal routing and excessive delays in horizontal communications. To address the above issues, this paper introduces a Connection-As-Required Scheme (CARS) specifically designed for delay-sensitive IoT and Cyber-Physical System (CPS) applications, where low-latency communication is essential for operational efficiency. CARS leverages Lyapunov optimization and backpressure algorithms to optimize traffic scheduling and routing, minimizing communication delay between entities. Benchmarking against state-of-the-art solutions, CARS reduces Round-Trip Time (RTT) to approximately 47.0% of conventional methods and decreases delay by 24.5% in TCP-based and 26.0% in UDP-based applications. These results highlight the potential of CARS to facilitate effective, low-latency collaboration in diverse I4.0 environments.</div></div>","PeriodicalId":48631,"journal":{"name":"Digital Communications and Networks","volume":"11 5","pages":"Pages 1519-1529"},"PeriodicalIF":7.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ExplainableDetector: Exploring transformer-based language modeling approach for SMS spam detection with explainability analysis ExplainableDetector：通过可解释性分析，探索基于转换器的SMS垃圾邮件检测语言建模方法

IF 7.5 2区计算机科学 Q1 TELECOMMUNICATIONS

Digital Communications and Networks

Pub Date : 2025-10-01 DOI: 10.1016/j.dcan.2025.07.008

Mohammad Amaz Uddin , Muhammad Nazrul Islam , Leandros Maglaras , Helge Janicke , Iqbal H. Sarker

Short Message Service (SMS) is a widely used and cost-effective communication medium that has unfortunately become a frequent target for unsolicited messages - commonly known as SMS spam. With the rapid adoption of smartphones and increased Internet connectivity, SMS spam has emerged as a prevalent threat. Spammers have recognized the critical role SMS plays in today's modern communication, making it a prime target for abuse. As cybersecurity threats continue to evolve, the volume of SMS spam has increased substantially in recent years. Moreover, the unstructured format of SMS data creates significant challenges for SMS spam detection, making it more difficult to successfully combat spam attacks. In this paper, we present an optimized and fine-tuned transformer-based Language Model to address the problem of SMS spam detection. We use a benchmark SMS spam dataset to analyze this spam detection model. Additionally, we utilize pre-processing techniques to obtain clean and noise-free data and address class imbalance problem by leveraging text augmentation techniques. The overall experiment showed that our optimized fine-tuned BERT (Bidirectional Encoder Representations from Transformers) variant model RoBERTa obtained high accuracy with 99.84%. To further enhance model transparency, we incorporate Explainable Artificial Intelligence (XAI) techniques that compute positive and negative coefficient scores, offering insight into the model's decision-making process. Additionally, we evaluate the performance of traditional machine learning models as a baseline for comparison. This comprehensive analysis demonstrates the significant impact language models can have on addressing complex text-based challenges within the cybersecurity landscape.

短消息服务（SMS）是一种广泛使用且具有成本效益的通信媒介，不幸的是，它已成为未经请求的消息（通常称为SMS垃圾邮件）的频繁目标。随着智能手机的迅速普及和互联网连接的增加，垃圾短信已经成为一种普遍的威胁。垃圾邮件发送者已经认识到SMS在当今现代通信中的关键作用，使其成为滥用的主要目标。随着网络安全威胁的不断发展，垃圾短信的数量近年来大幅增加。此外，SMS数据的非结构化格式为SMS垃圾邮件检测带来了重大挑战，使成功打击垃圾邮件攻击变得更加困难。在本文中，我们提出了一个优化和微调的基于转换的语言模型来解决短信垃圾邮件检测问题。我们使用基准短信垃圾邮件数据集来分析该垃圾邮件检测模型。此外，我们利用预处理技术获得干净和无噪声的数据，并利用文本增强技术解决类不平衡问题。整体实验表明，我们优化的微调BERT （Bidirectional Encoder Representations from Transformers）变体模型RoBERTa获得了99.84%的高精度。为了进一步提高模型的透明度，我们采用了可解释人工智能（XAI）技术来计算正系数和负系数分数，从而深入了解模型的决策过程。此外，我们评估传统机器学习模型的性能作为比较的基线。这一综合分析表明，语言模型对解决网络安全领域中基于文本的复杂挑战具有重要影响。

{"title":"ExplainableDetector: Exploring transformer-based language modeling approach for SMS spam detection with explainability analysis","authors":"Mohammad Amaz Uddin , Muhammad Nazrul Islam , Leandros Maglaras , Helge Janicke , Iqbal H. Sarker","doi":"10.1016/j.dcan.2025.07.008","DOIUrl":"10.1016/j.dcan.2025.07.008","url":null,"abstract":"<div><div>Short Message Service (SMS) is a widely used and cost-effective communication medium that has unfortunately become a frequent target for unsolicited messages - commonly known as SMS spam. With the rapid adoption of smartphones and increased Internet connectivity, SMS spam has emerged as a prevalent threat. Spammers have recognized the critical role SMS plays in today's modern communication, making it a prime target for abuse. As cybersecurity threats continue to evolve, the volume of SMS spam has increased substantially in recent years. Moreover, the unstructured format of SMS data creates significant challenges for SMS spam detection, making it more difficult to successfully combat spam attacks. In this paper, we present an optimized and fine-tuned transformer-based Language Model to address the problem of SMS spam detection. We use a benchmark SMS spam dataset to analyze this spam detection model. Additionally, we utilize pre-processing techniques to obtain clean and noise-free data and address class imbalance problem by leveraging text augmentation techniques. The overall experiment showed that our optimized fine-tuned BERT (Bidirectional Encoder Representations from Transformers) variant model RoBERTa obtained high accuracy with 99.84%. To further enhance model transparency, we incorporate Explainable Artificial Intelligence (XAI) techniques that compute positive and negative coefficient scores, offering insight into the model's decision-making process. Additionally, we evaluate the performance of traditional machine learning models as a baseline for comparison. This comprehensive analysis demonstrates the significant impact language models can have on addressing complex text-based challenges within the cybersecurity landscape.</div></div>","PeriodicalId":48631,"journal":{"name":"Digital Communications and Networks","volume":"11 5","pages":"Pages 1504-1518"},"PeriodicalIF":7.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529392","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Sensing accuracy gain, unified performance analysis and optimization in 6G cooperative ISAC systems 6G协同ISAC系统传感精度增益、统一性能分析与优化

IF 7.5 2区计算机科学 Q1 TELECOMMUNICATIONS

Digital Communications and Networks

Pub Date : 2025-10-01 DOI: 10.1016/j.dcan.2025.04.007

Guangyi Liu, Lincong Han, Rongyan Xi, Jing Dong, Jing Jin, Qixing Wang

Sixth Generation (6G) mobile communication networks will involve sensing as a new function, with the overwhelming trend of Integrated Sensing And Communications (ISAC). Although expanding the serving range of the networks, there exists performance trade-off between communication and sensing, in that they have competitions on the physical resources. Different resource allocation schemes will result in different sensing and communication performance, thus influencing the system's overall performance. Therefore, how to model the system's overall performance, and how to optimize it are key issues for ISAC. Relying on the large-scale deployment of the networks, cooperative ISAC has the advantages of wider coverage, more robust performance and good compatibility of multiple monostatic and multistatic sensing, compared to the non-cooperative ISAC. How to capture the performance gain of cooperation is a key issue for cooperative ISAC. To address the aforementioned vital problems, in this paper, we analyze the sensing accuracy gain, propose a unified ISAC performance evaluation framework and design several optimization methods in cooperative ISAC systems. The cooperative sensing accuracy gain is theoretically analyzed via Cramér Rao lower bound. The unified ISAC performance evaluation model is established by converting the communication mutual information to the effective minimum mean squared error. To optimize the unified ISAC performance, we design the optimization algorithms considering three factors: base stations' working modes, power allocation schemes and waveform design. Through simulations, we show the performance gain of the cooperative ISAC system and the effectiveness of the proposed optimization methods.

第六代（6G）移动通信网络将把传感作为一种新功能纳入其中，集成传感与通信（ISAC）是势不可挡的趋势。在扩大网络服务范围的同时，通信和感知之间存在着性能权衡，它们在物理资源上存在竞争。不同的资源分配方案会导致不同的感知和通信性能，从而影响系统的整体性能。因此，如何对系统的整体性能进行建模，并对其进行优化是ISAC面临的关键问题。依托于网络的大规模部署，合作ISAC相比非合作ISAC具有覆盖范围更广、性能更稳健、多单站和多站感知兼容性好等优点。如何捕捉合作的绩效增益是合作ISAC的关键问题。针对上述关键问题，本文分析了协同ISAC系统的传感精度增益，提出了统一的ISAC性能评估框架，并设计了几种优化方法。利用cramsamr - Rao下界理论分析了协同传感精度增益。将通信互信息转化为有效的最小均方误差，建立了统一的ISAC性能评价模型。为了优化统一ISAC性能，我们设计了考虑基站工作模式、功率分配方案和波形设计三个因素的优化算法。通过仿真，我们证明了协同ISAC系统的性能增益和所提出的优化方法的有效性。

{"title":"Sensing accuracy gain, unified performance analysis and optimization in 6G cooperative ISAC systems","authors":"Guangyi Liu, Lincong Han, Rongyan Xi, Jing Dong, Jing Jin, Qixing Wang","doi":"10.1016/j.dcan.2025.04.007","DOIUrl":"10.1016/j.dcan.2025.04.007","url":null,"abstract":"<div><div>Sixth Generation (6G) mobile communication networks will involve sensing as a new function, with the overwhelming trend of Integrated Sensing And Communications (ISAC). Although expanding the serving range of the networks, there exists performance trade-off between communication and sensing, in that they have competitions on the physical resources. Different resource allocation schemes will result in different sensing and communication performance, thus influencing the system's overall performance. Therefore, how to model the system's overall performance, and how to optimize it are key issues for ISAC. Relying on the large-scale deployment of the networks, cooperative ISAC has the advantages of wider coverage, more robust performance and good compatibility of multiple monostatic and multistatic sensing, compared to the non-cooperative ISAC. How to capture the performance gain of cooperation is a key issue for cooperative ISAC. To address the aforementioned vital problems, in this paper, we analyze the sensing accuracy gain, propose a unified ISAC performance evaluation framework and design several optimization methods in cooperative ISAC systems. The cooperative sensing accuracy gain is theoretically analyzed via Cramér Rao lower bound. The unified ISAC performance evaluation model is established by converting the communication mutual information to the effective minimum mean squared error. To optimize the unified ISAC performance, we design the optimization algorithms considering three factors: base stations' working modes, power allocation schemes and waveform design. Through simulations, we show the performance gain of the cooperative ISAC system and the effectiveness of the proposed optimization methods.</div></div>","PeriodicalId":48631,"journal":{"name":"Digital Communications and Networks","volume":"11 5","pages":"Pages 1657-1667"},"PeriodicalIF":7.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Distributed service caching with deep reinforcement learning for sustainable edge computing in large-scale AI 基于深度强化学习的分布式服务缓存在大规模人工智能中的可持续边缘计算

IF 7.5 2区计算机科学 Q1 TELECOMMUNICATIONS

Digital Communications and Networks

Pub Date : 2025-10-01 DOI: 10.1016/j.dcan.2024.11.009

Wei Liu , Muhammad Bilal , Yuzhe Shi , Xiaolong Xu

Increasing reliance on large-scale AI models has led to rising demand for intelligent services. The centralized cloud computing approach has limitations in terms of data transfer efficiency and response time, and as a result many service providers have begun to deploy edge servers to cache intelligent services in order to reduce transmission delay and communication energy consumption. However, finding the optimal service caching strategy remains a significant challenge due to the stochastic nature of service requests and the bulky nature of intelligent services. To deal with this, we propose a distributed service caching scheme integrating deep reinforcement learning (DRL) with mobility prediction, which we refer to as DSDM. Specifically, we employ the D3QN (Deep Double Dueling Q-Network) framework to integrate Long Short-Term Memory (LSTM) predicted mobile device locations into the service caching replacement algorithm and adopt the distributed multi-agent approach for learning and training. Experimental results demonstrate that DSDM achieves significant performance improvements in reducing communication energy consumption compared to traditional methods across various scenarios.

越来越多地依赖大规模人工智能模型，导致对智能服务的需求不断上升。集中式云计算方法在数据传输效率和响应时间方面存在局限性，因此许多服务提供商已开始部署边缘服务器来缓存智能服务，以减少传输延迟和通信能耗。然而，由于服务请求的随机性和智能服务的庞大性，找到最优的服务缓存策略仍然是一个重大挑战。为了解决这个问题，我们提出了一种集成深度强化学习（DRL）和移动性预测的分布式服务缓存方案，我们称之为DSDM。具体而言，我们采用D3QN （Deep Double Dueling Q-Network）框架将长短期记忆（LSTM）预测的移动设备位置集成到服务缓存替换算法中，并采用分布式多智能体方法进行学习和训练。实验结果表明，在各种场景下，与传统方法相比，DSDM在降低通信能耗方面取得了显著的性能提升。

{"title":"Distributed service caching with deep reinforcement learning for sustainable edge computing in large-scale AI","authors":"Wei Liu , Muhammad Bilal , Yuzhe Shi , Xiaolong Xu","doi":"10.1016/j.dcan.2024.11.009","DOIUrl":"10.1016/j.dcan.2024.11.009","url":null,"abstract":"<div><div>Increasing reliance on large-scale AI models has led to rising demand for intelligent services. The centralized cloud computing approach has limitations in terms of data transfer efficiency and response time, and as a result many service providers have begun to deploy edge servers to cache intelligent services in order to reduce transmission delay and communication energy consumption. However, finding the optimal service caching strategy remains a significant challenge due to the stochastic nature of service requests and the bulky nature of intelligent services. To deal with this, we propose a distributed service caching scheme integrating deep reinforcement learning (DRL) with mobility prediction, which we refer to as DSDM. Specifically, we employ the D3QN (Deep Double Dueling Q-Network) framework to integrate Long Short-Term Memory (LSTM) predicted mobile device locations into the service caching replacement algorithm and adopt the distributed multi-agent approach for learning and training. Experimental results demonstrate that DSDM achieves significant performance improvements in reducing communication energy consumption compared to traditional methods across various scenarios.</div></div>","PeriodicalId":48631,"journal":{"name":"Digital Communications and Networks","volume":"11 5","pages":"Pages 1447-1456"},"PeriodicalIF":7.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Standardised interworking and deployment of IoT and edge computing platforms 物联网和边缘计算平台的标准化互通和部署

IF 7.5 2区计算机科学 Q1 TELECOMMUNICATIONS

Digital Communications and Networks

Pub Date : 2025-10-01 DOI: 10.1016/j.dcan.2025.04.006

Jieun Lee , JooSung Kim , Seong Ki Yoo , Tarik Taleb , JaeSeung Song

Edge computing is swiftly gaining traction and is being standardised by the European Telecommunications Standards Institute (ETSI) as Multi-access Edge Computing (MEC). Simultaneously, oneM2M has been actively developing standards for dynamic data management and IoT services at the edge, particularly for applications that require real-time support and security. Integrating MEC and oneM2M offers a unique opportunity to maximize their individual strengths. Therefore, this article proposes a framework that integrates MEC and oneM2M standard platforms for IoT applications, demonstrating how the synergy of these architectures can leverage the geographically distributed computing resources at base stations, enabling efficient deployment and added value for time-sensitive IoT applications. In addition, this study offers a concept of potential interworking models between oneM2M and the MEC architectures. The adoption of these standard architectures can enable various IoT edge services, such as smart city mobility and real-time analytics functions, by leveraging the oneM2M common service layer instantiated on the MEC host.

边缘计算正迅速获得关注，并被欧洲电信标准协会（ETSI）标准化为多接入边缘计算（MEC）。同时，oneM2M一直在积极开发边缘动态数据管理和物联网服务标准，特别是需要实时支持和安全性的应用。整合MEC和oneM2M提供了一个独特的机会，最大限度地发挥各自的优势。因此，本文提出了一个集成MEC和oneM2M标准平台的物联网应用框架，展示了这些架构的协同作用如何利用基站的地理分布式计算资源，为时间敏感的物联网应用实现高效部署和增值。此外，本研究提供了oneM2M和MEC架构之间潜在的交互模型的概念。通过利用在MEC主机上实例化的oneM2M公共服务层，采用这些标准架构可以实现各种物联网边缘服务，例如智慧城市移动性和实时分析功能。

{"title":"Standardised interworking and deployment of IoT and edge computing platforms","authors":"Jieun Lee , JooSung Kim , Seong Ki Yoo , Tarik Taleb , JaeSeung Song","doi":"10.1016/j.dcan.2025.04.006","DOIUrl":"10.1016/j.dcan.2025.04.006","url":null,"abstract":"<div><div>Edge computing is swiftly gaining traction and is being standardised by the European Telecommunications Standards Institute (ETSI) as Multi-access Edge Computing (MEC). Simultaneously, oneM2M has been actively developing standards for dynamic data management and IoT services at the edge, particularly for applications that require real-time support and security. Integrating MEC and oneM2M offers a unique opportunity to maximize their individual strengths. Therefore, this article proposes a framework that integrates MEC and oneM2M standard platforms for IoT applications, demonstrating how the synergy of these architectures can leverage the geographically distributed computing resources at base stations, enabling efficient deployment and added value for time-sensitive IoT applications. In addition, this study offers a concept of potential interworking models between oneM2M and the MEC architectures. The adoption of these standard architectures can enable various IoT edge services, such as smart city mobility and real-time analytics functions, by leveraging the oneM2M common service layer instantiated on the MEC host.</div></div>","PeriodicalId":48631,"journal":{"name":"Digital Communications and Networks","volume":"11 5","pages":"Pages 1578-1587"},"PeriodicalIF":7.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cross-feature fusion speech emotion recognition based on attention mask residual network and Wav2vec 2.0 基于注意掩模残差网络和Wav2vec 2.0的跨特征融合语音情感识别

IF 7.5 2区计算机科学 Q1 TELECOMMUNICATIONS

Digital Communications and Networks

Pub Date : 2025-10-01 DOI: 10.1016/j.dcan.2024.10.007

Xiaoke Li, Zufan Zhang

Speech Emotion Recognition (SER) has received widespread attention as a crucial way for understanding human emotional states. However, the impact of irrelevant information on speech signals and data sparsity limit the development of SER system. To address these issues, this paper proposes a framework that incorporates the Attentive Mask Residual Network (AM-ResNet) and the self-supervised learning model Wav2vec 2.0 to obtain AM-ResNet features and Wav2vec 2.0 features respectively, together with a cross-attention module to interact and fuse these two features. The AM-ResNet branch mainly consists of maximum amplitude difference detection, mask residual block, and an attention mechanism. Among them, the maximum amplitude difference detection and the mask residual block act on the pre-processing and the network, respectively, to reduce the impact of silent frames, and the attention mechanism assigns different weights to unvoiced and voiced speech to reduce redundant emotional information caused by unvoiced speech. In the Wav2vec 2.0 branch, this model is introduced as a feature extractor to obtain general speech features (Wav2vec 2.0 features) through pre-training with a large amount of unlabeled speech data, which can assist the SER task and cope with data sparsity problems. In the cross-attention module, AM-ResNet features and Wav2vec 2.0 features are interacted with and fused to obtain the cross-fused features, which are used to predict the final emotion. Furthermore, multi-label learning is also used to add ambiguous emotion utterances to deal with data limitations. Finally, experimental results illustrate the usefulness and superiority of our proposed framework over existing state-of-the-art approaches.

语音情绪识别（SER）作为理解人类情绪状态的重要手段，受到了广泛的关注。然而，语音信号中不相关信息的影响和数据的稀疏性限制了语音识别系统的发展。为了解决这些问题，本文提出了一个框架，该框架结合了关注遮罩残差网络（AM-ResNet）和自监督学习模型Wav2vec 2.0，分别获得AM-ResNet特征和Wav2vec 2.0特征，并使用交叉注意模块来交互和融合这两个特征。AM-ResNet分支主要由最大幅差检测、掩码残差块和注意机制组成。其中，最大振幅差检测和掩模残差块分别作用于预处理和网络，以减少无声帧的影响，注意机制对不发音和不发音语音赋予不同的权重，以减少不发音语音带来的冗余情感信息。在Wav2vec 2.0分支中，引入了该模型作为特征提取器，通过对大量未标记语音数据进行预训练，获得一般语音特征（Wav2vec 2.0特征），可以辅助SER任务，解决数据稀疏性问题。在交叉关注模块中，AM-ResNet特征与Wav2vec 2.0特征进行交互融合，得到交叉融合特征，用于预测最终情绪。此外，多标签学习还用于添加模棱两可的情感话语，以解决数据限制问题。最后，实验结果表明我们提出的框架比现有的最先进的方法有用和优越。

{"title":"Cross-feature fusion speech emotion recognition based on attention mask residual network and Wav2vec 2.0","authors":"Xiaoke Li, Zufan Zhang","doi":"10.1016/j.dcan.2024.10.007","DOIUrl":"10.1016/j.dcan.2024.10.007","url":null,"abstract":"<div><div>Speech Emotion Recognition (SER) has received widespread attention as a crucial way for understanding human emotional states. However, the impact of irrelevant information on speech signals and data sparsity limit the development of SER system. To address these issues, this paper proposes a framework that incorporates the Attentive Mask Residual Network (AM-ResNet) and the self-supervised learning model Wav2vec 2.0 to obtain AM-ResNet features and Wav2vec 2.0 features respectively, together with a cross-attention module to interact and fuse these two features. The AM-ResNet branch mainly consists of maximum amplitude difference detection, mask residual block, and an attention mechanism. Among them, the maximum amplitude difference detection and the mask residual block act on the pre-processing and the network, respectively, to reduce the impact of silent frames, and the attention mechanism assigns different weights to unvoiced and voiced speech to reduce redundant emotional information caused by unvoiced speech. In the Wav2vec 2.0 branch, this model is introduced as a feature extractor to obtain general speech features (Wav2vec 2.0 features) through pre-training with a large amount of unlabeled speech data, which can assist the SER task and cope with data sparsity problems. In the cross-attention module, AM-ResNet features and Wav2vec 2.0 features are interacted with and fused to obtain the cross-fused features, which are used to predict the final emotion. Furthermore, multi-label learning is also used to add ambiguous emotion utterances to deal with data limitations. Finally, experimental results illustrate the usefulness and superiority of our proposed framework over existing state-of-the-art approaches.</div></div>","PeriodicalId":48631,"journal":{"name":"Digital Communications and Networks","volume":"11 5","pages":"Pages 1567-1577"},"PeriodicalIF":7.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Maximizing energy efficiency in 6G cognitive radio network 最大化6G认知无线网络的能源效率

IF 7.5 2区计算机科学 Q1 TELECOMMUNICATIONS

Digital Communications and Networks

Pub Date : 2025-10-01 DOI: 10.1016/j.dcan.2025.06.008

Umar Ghafoor, Adil Masood Siddiqui

The increasing demand for infotainment applications necessitates efficient bandwidth and energy resource allocation. Sixth-Generation (6G) networks, utilizing Cognitive Radio (CR) technology within CR Network (CRN), can enhance spectrum utilization by accessing unused spectrum when licensed Primary Mobile Equipment (PME) is inactive or served by a Primary Base Station (PrBS). Secondary Mobile Equipment (SME) accesses this spectrum through a Secondary Base Station (SrBS) using opportunistic access, i.e., spectrum sensing. Hybrid Multiple Access (HMA), combining Orthogonal Multiple Access (OMA) and Non-Orthogonal Multiple Access (NOMA), can enhance Energy Efficiency (EE). Additionally, SME Clustering (SMEC) reduces inter-cluster interference, enhancing EE further. Despite these advancements, the integration of CR technology, HMA, and SMEC in CRN for better bandwidth utilization and EE remains unexplored. This paper introduces a new CR-assisted SMEC-based Downlink HMA (CR-SMEC-DHMA) method for 6G CRN, aimed at jointly optimizing SME admission, SME association, sum rate, and EE subject to imperfect sensing, collision, and Quality of Service (QoS). A novel optimization problem, formulated as a non-linear fractional programming problem, is solved using the Charnes-Cooper Transformation (CCT) to convert into a concave optimization problem, and an ϵ-optimal Outer Approximation Algorithm (OAA) is employed to solve the concave optimization problem. Simulations demonstrate the effectiveness of the proposed CR-SMEC-DHMA, surpassing the performance of current OMA-enabled CRN, NOMA-enabled CRN, SMEC-OMA enabled CRN, and SMEC-NOMA enabled CRN methods, with ϵ-optimal results obtained at

ϵ = 10^{- 3}

, while satisfying Performance Measures (PMs) including SME admission in SMEC, SME association with SrBS, SME-channel opportunistic allocation through spectrum sensing, sum rate and overall EE within the 6G CRN.

对信息娱乐应用日益增长的需求需要有效的带宽和能源分配。第六代（6G）网络利用认知无线电（CR）网络（CRN）中的认知无线电（CR）技术，当授权的主要移动设备（PME）处于非活动状态或由主要基站（PrBS）提供服务时，可以通过访问未使用的频谱来提高频谱利用率。二次移动设备（SME）通过二次基站（SrBS）使用机会接入，即频谱感知，访问该频谱。混合多址（HMA）是正交多址（OMA）和非正交多址（NOMA）的结合，可以提高能源效率（EE）。此外，中小企业集群（SMEC）减少了集群间的干扰，进一步提高了EE。尽管取得了这些进步，但在CRN中集成CR技术、HMA和SMEC以获得更好的带宽利用率和EE仍有待探索。本文提出了一种新的基于cr辅助smec的6G CRN下行HMA （CR-SMEC-DHMA）方法，旨在共同优化感知、碰撞和服务质量（QoS）不完美情况下的中小企业准入、中小企业关联、和率和EE。利用Charnes-Cooper变换（CCT）将非线性分式规划问题转化为凹优化问题，并利用ϵ-optimal外逼近算法（OAA）求解凹优化问题。仿真证明了所提出的CR-SMEC-DHMA的有效性，超越了当前支持oma的CRN、支持noma的CRN、支持SMEC- oma的CRN和支持SMEC- noma的CRN方法的性能，ϵ-optimal在λ =10−3处获得的结果，同时满足性能指标（PMs），包括SMEC中中小企业的准入、中小企业与SrBS的关联、通过频谱感知的中小企业通道机会分配、求和速率和6G CRN内的总体EE。

{"title":"Maximizing energy efficiency in 6G cognitive radio network","authors":"Umar Ghafoor, Adil Masood Siddiqui","doi":"10.1016/j.dcan.2025.06.008","DOIUrl":"10.1016/j.dcan.2025.06.008","url":null,"abstract":"<div><div>The increasing demand for infotainment applications necessitates efficient bandwidth and energy resource allocation. Sixth-Generation (6G) networks, utilizing Cognitive Radio (CR) technology within CR Network (CRN), can enhance spectrum utilization by accessing unused spectrum when licensed Primary Mobile Equipment (PME) is inactive or served by a Primary Base Station (PrBS). Secondary Mobile Equipment (SME) accesses this spectrum through a Secondary Base Station (SrBS) using opportunistic access, i.e., spectrum sensing. Hybrid Multiple Access (HMA), combining Orthogonal Multiple Access (OMA) and Non-Orthogonal Multiple Access (NOMA), can enhance Energy Efficiency (EE). Additionally, SME Clustering (SMEC) reduces inter-cluster interference, enhancing EE further. Despite these advancements, the integration of CR technology, HMA, and SMEC in CRN for better bandwidth utilization and EE remains unexplored. This paper introduces a new CR-assisted SMEC-based Downlink HMA (CR-SMEC-DHMA) method for 6G CRN, aimed at jointly optimizing SME admission, SME association, sum rate, and EE subject to imperfect sensing, collision, and Quality of Service (QoS). A novel optimization problem, formulated as a non-linear fractional programming problem, is solved using the Charnes-Cooper Transformation (CCT) to convert into a concave optimization problem, and an <em>ϵ</em>-optimal Outer Approximation Algorithm (OAA) is employed to solve the concave optimization problem. Simulations demonstrate the effectiveness of the proposed CR-SMEC-DHMA, surpassing the performance of current OMA-enabled CRN, NOMA-enabled CRN, SMEC-OMA enabled CRN, and SMEC-NOMA enabled CRN methods, with <em>ϵ</em>-optimal results obtained at <span><math><mi>ϵ</mi><mo>=</mo><msup><mrow><mn>10</mn></mrow><mrow><mo>−</mo><mn>3</mn></mrow></msup></math></span>, while satisfying Performance Measures (PMs) including SME admission in SMEC, SME association with SrBS, SME-channel opportunistic allocation through spectrum sensing, sum rate and overall EE within the 6G CRN.</div></div>","PeriodicalId":48631,"journal":{"name":"Digital Communications and Networks","volume":"11 5","pages":"Pages 1356-1369"},"PeriodicalIF":7.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Chinese relation extraction for constructing satellite frequency and orbit knowledge graph: A survey 构建卫星频率与轨道知识图谱的中文关系提取综述

IF 7.5 2区计算机科学 Q1 TELECOMMUNICATIONS

Digital Communications and Networks

Pub Date : 2025-10-01 DOI: 10.1016/j.dcan.2025.05.002

Yuanzhi He , Zhiqiang Li , Zheng Dou

As Satellite Frequency and Orbit (SFO) constitute scarce natural resources, constructing a Satellite Frequency and Orbit Knowledge Graph (SFO-KG) becomes crucial for optimizing their utilization. In the process of building the SFO-KG from Chinese unstructured data, extracting Chinese entity relations is the fundamental step. Although Relation Extraction (RE) methods in the English field have been extensively studied and developed earlier than their Chinese counterparts, their direct application to Chinese texts faces significant challenges due to linguistic distinctions such as unique grammar, pictographic characters, and prevalent polysemy. The absence of comprehensive reviews on Chinese RE research progress necessitates a systematic investigation. A thorough review of Chinese RE has been conducted from four methodological approaches: pipeline RE, joint entity-relation extraction, open domain RE, and multimodal RE techniques. In addition, we further analyze the essential research infrastructure, including specialized datasets, evaluation benchmarks, and competitions within Chinese RE research. Finally, the current research challenges and development trends in the field of Chinese RE were summarized and analyzed from the perspectives of ecological construction methods for datasets, open domain RE, N-ary RE, and RE based on large language models. This comprehensive review aims to facilitate SFO-KG construction and its practical applications in SFO resource management.

卫星频率与轨道是稀缺的自然资源，构建卫星频率与轨道知识图谱是优化卫星频率与轨道资源利用的关键。在中文非结构化数据构建SFO-KG的过程中，中文实体关系的提取是基础步骤。尽管关系提取方法在英语领域的研究和发展要早于汉语，但由于其独特的语法、象形文字和普遍存在的一词多义等语言差异，关系提取方法在汉语文本中的直接应用面临着重大挑战。缺乏对中国可再生能源研究进展的全面综述，有必要进行系统的调查。本文从管道可再生能源、联合实体关系提取、开放域可再生能源和多模态可再生能源技术四种方法对中国可再生能源进行了全面的综述。此外，我们进一步分析了中国可再生能源研究的基本研究基础设施，包括专业数据集、评估基准和竞争。最后，从数据集生态构建方法、开放域RE、N-ary RE和基于大语言模型的RE等方面，总结和分析了当前中文RE领域的研究挑战和发展趋势。本文综述旨在促进SFO- kg的构建及其在SFO资源管理中的实际应用。

{"title":"Chinese relation extraction for constructing satellite frequency and orbit knowledge graph: A survey","authors":"Yuanzhi He , Zhiqiang Li , Zheng Dou","doi":"10.1016/j.dcan.2025.05.002","DOIUrl":"10.1016/j.dcan.2025.05.002","url":null,"abstract":"<div><div>As Satellite Frequency and Orbit (SFO) constitute scarce natural resources, constructing a Satellite Frequency and Orbit Knowledge Graph (SFO-KG) becomes crucial for optimizing their utilization. In the process of building the SFO-KG from Chinese unstructured data, extracting Chinese entity relations is the fundamental step. Although Relation Extraction (RE) methods in the English field have been extensively studied and developed earlier than their Chinese counterparts, their direct application to Chinese texts faces significant challenges due to linguistic distinctions such as unique grammar, pictographic characters, and prevalent polysemy. The absence of comprehensive reviews on Chinese RE research progress necessitates a systematic investigation. A thorough review of Chinese RE has been conducted from four methodological approaches: pipeline RE, joint entity-relation extraction, open domain RE, and multimodal RE techniques. In addition, we further analyze the essential research infrastructure, including specialized datasets, evaluation benchmarks, and competitions within Chinese RE research. Finally, the current research challenges and development trends in the field of Chinese RE were summarized and analyzed from the perspectives of ecological construction methods for datasets, open domain RE, N-ary RE, and RE based on large language models. This comprehensive review aims to facilitate SFO-KG construction and its practical applications in SFO resource management.</div></div>","PeriodicalId":48631,"journal":{"name":"Digital Communications and Networks","volume":"11 5","pages":"Pages 1305-1317"},"PeriodicalIF":7.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529501","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0